US9081663B2 - Optimized garbage collection algorithm to improve solid state drive reliability - Google Patents

Optimized garbage collection algorithm to improve solid state drive reliability Download PDF

Info

Publication number
US9081663B2
US9081663B2 US13/664,335 US201213664335A US9081663B2 US 9081663 B2 US9081663 B2 US 9081663B2 US 201213664335 A US201213664335 A US 201213664335A US 9081663 B2 US9081663 B2 US 9081663B2
Authority
US
United States
Prior art keywords
plurality
data blocks
rank
number
associated
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active, expires
Application number
US13/664,335
Other versions
US20130132647A1 (en
Inventor
Ashot Melik-Martirosian
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Western Digital Technologies Inc
Original Assignee
HGST Technologies Santa Ana Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority to US201161561663P priority Critical
Application filed by HGST Technologies Santa Ana Inc filed Critical HGST Technologies Santa Ana Inc
Priority to US13/664,335 priority patent/US9081663B2/en
Publication of US20130132647A1 publication Critical patent/US20130132647A1/en
Assigned to STEC, INC. reassignment STEC, INC. ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: MELIK-MARTIROSIAN, ASHOT
Assigned to HGST TECHNOLOGIES SANTA ANA, INC. reassignment HGST TECHNOLOGIES SANTA ANA, INC. CHANGE OF NAME (SEE DOCUMENT FOR DETAILS). Assignors: STEC, INC.
Publication of US9081663B2 publication Critical patent/US9081663B2/en
Application granted granted Critical
Assigned to WESTERN DIGITAL TECHNOLOGIES, INC. reassignment WESTERN DIGITAL TECHNOLOGIES, INC. ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: HGST TECHNOLOGIES SANTA ANA, INC.
Application status is Active legal-status Critical
Adjusted expiration legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02Addressing or allocation; Relocation
    • G06F12/0223User address space allocation, e.g. contiguous or non contiguous base addressing
    • G06F12/023Free address space management
    • G06F12/0253Garbage collection, i.e. reclamation of unreferenced memory
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02Addressing or allocation; Relocation
    • G06F12/0223User address space allocation, e.g. contiguous or non contiguous base addressing
    • G06F12/023Free address space management
    • G06F12/0238Memory management in non-volatile memory, e.g. resistive RAM or ferroelectric memory
    • G06F12/0246Memory management in non-volatile memory, e.g. resistive RAM or ferroelectric memory in block erasable memory, e.g. flash memory
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06Digital input from or digital output to record carriers, e.g. RAID, emulated record carriers, networked record carriers
    • G06F3/0601Dedicated interfaces to storage systems
    • G06F3/0602Dedicated interfaces to storage systems specifically adapted to achieve a particular effect
    • G06F3/0626Reducing size or complexity of storage systems
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06Digital input from or digital output to record carriers, e.g. RAID, emulated record carriers, networked record carriers
    • G06F3/0601Dedicated interfaces to storage systems
    • G06F3/0628Dedicated interfaces to storage systems making use of a particular technique
    • G06F3/0638Organizing or formatting or addressing of data
    • G06F3/0644Management of space entities, e.g. partitions, extents, pools
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06Digital input from or digital output to record carriers, e.g. RAID, emulated record carriers, networked record carriers
    • G06F3/0601Dedicated interfaces to storage systems
    • G06F3/0668Dedicated interfaces to storage systems adopting a particular infrastructure
    • G06F3/0671In-line storage system
    • G06F3/0683Plurality of storage devices
    • G06F3/0689Disk arrays, e.g. RAID, JBOD
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F2212/00Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
    • G06F2212/72Details relating to flash memory management
    • G06F2212/7205Cleaning, compaction, garbage collection, erase control
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F2212/00Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
    • G06F2212/72Details relating to flash memory management
    • G06F2212/7211Wear leveling

Abstract

A method for managing memory operations in a storage device having a plurality of data blocks, the method including steps for determining a number of invalid pages, in each of the plurality of data blocks, determining a number of page reads for each of the plurality of data blocks and determining a dwell time for each of the plurality of data blocks. In certain aspects, the method further comprises steps for selecting a data block, from among the plurality of data blocks, for memory reclamation based on the number of invalid pages, the number of page reads, and the dwell time of the selected data block. A flash storage system and computer-readable media are also provided.

Description

This application claims the benefit of U.S. Provisional Application No. 61/561,663, entitled “OPTIMIZED GARBAGE COLLECTION ALGORITHM TO IMPROVE SSD RELIABILITY,” filed Nov. 18, 2011, which is incorporated herein by reference.

BACKGROUND

The subject disclosure relates to solid state memory devices, such as solid-state drives (SSDs). Particularly, the subject disclosure pertains to improved memory reclamation algorithms.

SUMMARY

Although conventional reclaiming algorithms are relatively simple, they typically do not account for memory reliability considerations, which are becoming more relevant in scaled memory devices. In particular, conventional reclaiming algorithms do not take into consideration the limited read disturb and retention characteristics of flash devices, such as, MLC NAND flash devices frequently used in enterprise solid state drive (SSD) applications. The subject technology provides an improved memory reclaiming algorithm (e.g., a “garbage collection algorithm”) that overcomes some of the limitations of the existing reclaiming algorithms by taking into consideration reliability limitations. More specifically, aspects of the subject technology take into consideration the limited retention and read disturb characteristics of memory devices, such as MLC NAND Flash, while still maintaining an efficient garbage collection capability.

In certain aspects, the subject technology relates to a method for managing memory operations in a storage device having a plurality of data blocks, the method includes steps for determining a number of invalid pages in each of the plurality of data blocks, determining a number of page reads for each of the plurality of data blocks, and determining a dwell time for each of the plurality of data blocks. In certain aspects, the method further includes steps for selecting a data block, from among the plurality of data blocks, for memory reclamation based on the number of invalid pages, the number of page reads, and the dwell time of the selected data block.

In another aspect, the subject technology relates to a flash storage system comprising a memory, a flash memory array comprising a plurality of flash memory blocks and a controller coupled to the memory and the flash memory array, wherein the controller is configured to perform operations for storing to the memory, a number of invalid pages associated with each of the plurality of data blocks, storing to the memory, a number of page reads associated with each of the plurality of data blocks and storing to the memory, a dwell time associated with each of the plurality of data blocks. In certain aspects, the controller may be further configured to perform operations for selecting a data block, from among the plurality of data blocks, for memory reclamation based on the number of invalid pages, the number of page reads, and the dwell time of the selected data block.

In yet another aspect, the subject technology relates to a computer-readable storage medium comprising instructions stored therein, which when executed by a processor, cause the processor to perform operations comprising determining a number of invalid pages in each of the plurality of data blocks, determining a number of page reads for each of the plurality of data blocks and determining a dwell time for each of the plurality of data blocks. In certain aspects, the processor may further perform operations for selecting a data block, from among the plurality of data blocks, for memory reclamation based on the number of invalid pages, the number of page reads, and the dwell time of the selected data block.

It is understood that other configurations of the subject technology will become readily apparent to those skilled in the art from the following detailed description, wherein various configurations of the subject technology are shown and described by way of illustration. As will be realized, the subject technology is capable of other and different configurations and its several details are capable of modification in various other respects, all without departing from the scope of the subject technology. Accordingly, the drawings and detailed description are to be regarded as illustrative in nature and not as restrictive.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagram illustrating components of a flash memory device according to one aspect of the subject technology.

FIGS. 2A, 2B and 2C illustrate examples of a process to erase memory blocks that one or more pages of data to be modified.

FIG. 3 illustrates an example of the hierarchy of rank group (RG) levels according to some aspects of the subject disclosure.

FIG. 4 illustrates an example of the hierarchy of RG levels according to some aspects of the subject disclosure.

FIG. 5 illustrates an example of a write procedure performed to produce erased blocks out of blocks containing garbage data.

DETAILED DESCRIPTION

The detailed description set forth below is intended as a description of various configurations of the subject technology and is not intended to represent the only configurations in which the subject technology may be practiced. The appended drawings are incorporated herein and constitute a part of the detailed description. The detailed description includes specific details for the purpose of providing a thorough understanding of the subject technology. However, the subject technology may be practiced without these specific details. In some instances, structures and components are shown in block diagram form in order to avoid obscuring the concepts of the subject technology. Like components are labeled with identical element numbers for ease of understanding.

FIG. 1 is a block diagram illustrating components of flash memory device 100 according to one aspect of the subject technology. As depicted in FIG. 1, host 110 is coupled to flash memory device 100 via host interface 150. Flash memory device 100 includes controller 120 and flash memory array 130. Depending on implementation, the elements of flash memory device 100 can be integrated into a single chip or implemented in two or more discrete components.

Controller 120 may be implemented with a general-purpose microprocessor, a microcontroller, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), a programmable logic device (PLD), a controller, a state machine, gated logic, discrete hardware components, or a combination of the foregoing. One or more sequences of instructions may be stored as firmware on a ROM within controller 120. One or more sequences of instructions also may be software stored and read from another storage medium, such as flash memory array 130, or received from a host device (e.g., host 110) via host interface 150. ROMs, storage media, and flash memory arrays represent examples of machine or computer readable media that can store instructions/code executable by the controller. Machine or computer readable media may generally refer to any medium or media used to provide instructions to controller 120, including both volatile media, such as dynamic memory used for storage media or for buffers within controller 120, and non-volatile media, such as electronic media, optical media, and magnetic media.

Host interface 150 may be configured to implement a standard interface, such as Serial-Attached SCSI (SAS), Fiber Channel interface, PCI Express (PCIe), SATA, USB, and the like. Host interface 140 may be configured to implement only one interface. Alternatively, host interface 140 may be configured to implement multiple interfaces, which are individually selectable using a configuration parameter selected by a user or programmed at the time of assembly. Host interface 140 can include one or more buffers for buffering transmissions between a host device and the controller. A host device (e.g., host 110) may be any device configured to be coupled to the data storage system and to store data in data storage system. In some examples, the host device may be a computing system such as a personal computer, a server, a workstation, a laptop computer, PDA, smart phone, and the like. Alternatively, the host device may be an electronic device such as a digital camera, a digital audio player, a digital video recorder or the like.

Flash memory array 130 represents non-volatile memory devices for storing data. By way of example, flash memory array 130 may comprise a single-level cell (SLC) memory, a multi-level cell (MLC) memory or a three-level cell (TLC) memory device. In some aspects, flash memory array 130 may comprise one or more hybrid memory devices that can function in one or more of a SLC, MLC or TLC mode.

Each component of flash memory array 130 can include a single flash memory device or chip, or may include multiple flash memory devices or chips arranged in multiple channels, as depicted in FIG. 1. However, flash memory array 130 is not limited to any particular capacity or configuration.

The flash memory is organized into memory blocks, for example, depending on the flash memory generation. Memory blocks may include different numbers of pages, each of which may have different sizes. Although the number of cell states, the number of memory blocks, the number of physical pages per physical block, the number of sectors per physical page, and the size of the sectors can vary depending on implementation, in some examples, each memory block may consist of 64, 128 or 256 pages, etc. Furthermore, each page may vary in size, for example, 4 kB, 8 kB or 16 kB, etc.

FIG. 2A conceptually illustrates an example process for designating content that is to be modified as invalid, or “garbage” data. Specifically, FIG. 2A illustrates memory blocks 201-204, containing differently marked data portions. For example, memory blocks 201A-204A each include one or more pages of valid data (indicated as “V”), as well as one or more pages of data to be modified (indicated “to be modified”). In certain aspects, portions of memory blocks 201A-204A containing data that is to be modified are marked as “garbage,” as indicated by corresponding memory blocks 201B-204B. As will be discussed in further detail below, valid data of memory blocks 201B-204B can then be consolidated in a garbage collection procedure in order to produce available erased blocks.

FIG. 2B illustrates available erased blocks 205A-207A that do not contain garbage data and that are available to be written. For example, available erased blocks 205A-207A are available to be written with new content (e.g., valid data), as illustrated by memory blocks 205B-207B.

FIG. 2C further illustrates a garbage collection process, which produces B erased blocks out of X partially valid memory blocks. Specifically, FIG. 2C illustrates partially valid memory blocks 208A, 209A and 210A, where partially valid memory block 210A represents the Xth memory block in a series of X total partially valid memory blocks. FIG. 2C further illustrates available erased memory blocks 208B, 209B and 210B, wherein available erased memory block 210B represents the Ath memory block in a series of A total available erased memory blocks. Similarly, erased memory blocks 208C, 209C and 210C represent previously erased memory blocks now containing valid data, wherein erased memory block 210C represents the Bth memory block in a series of B total erased memory blocks.

In certain implementations, a total number of B erased blocks produced by the garbage collection process is given by:
B=Σ 1 X(V i)/N  (1)
where Vi is the number of valid pages in the ith block and N is the total number of pages in the block. In this example, A blocks are reclaimed where A is given by:
A=Σ 1 X(G i)/N  (2)
where Gi represents the number of garbage pages in the ith block. Therefore, for every A blocks to be written (e.g., by a host such as host 110), a total of A+B blocks must be written to flash memory, such as flash memory array 130. The ratio of blocks written internally to blocks written by the host is commonly referred to as “write amplification” (WA). In certain aspects, the WA can be given by:
WA=(A+B)/A=1+B/A=1+Σ1 X(V i)/Σ1 X(G i)=1+V ave /G ave  (3)
where Vave and Gave represent average numbers of valid pages and garbage pages across blocks, respectively. Because extra write operations can reduce write performance and consume the limited number of program/erase (P/E) cycles that every memory block can undergo, in certain aspects it is desirable to achieve a WA value as close to 1 as possible, for example, to avoid unnecessary internal write operations for every host operation. In some implementations, it can be difficult to achieve a WA value close to 1. Thus, in certain implementations WA may be larger.

By way of example, WA may range between 2.5 and 4 (with a Vave/Gave in the 1.5 to 2.0 range). Based on formula (3) given above, in order to reduce the WA, it is advantageous to first recycle (e.g., via garbage collection) blocks with the smallest Vave values. Thus, blocks with the lowest Vi and the highest time (since last write) are recycled first. In some implementations, the rationale behind time-indexing the memory blocks is that with the passing of time the host is likely to declare more pages in a given block as “garbage,” since new data is constantly written into the SSD.

Although this process accounts for a number of valid pages on a memory block and the time since a block was last written, the process does not take consideration of reliability concerns. In particular, the process does not consider the limited read disturb and retention characteristics of some flash memory devices, for example, MLC NAND flash devices.

Aspects of the subject technology provide an improved garbage collection procedure by taking into consideration reliability concerns through the indexing and ranking of memory blocks. In some implementations, the subject technology provides a method for freeing and consolidating memory in a manner that is based on memory block retention (e.g., data retention) and page read limitations.

As used herein “data retention” refers to the ability of a memory device (e.g., flash memory array 130) to maintain data after a specified time period, at a specified temperature. To achieve higher endurance (e.g., a greater number of P/E cycles), retention should be higher than the combined maximum time that data is stored in a power-down state (e.g., for three months at 40° C.) and the maximum time between memory block writes during memory operation, referred to herein as “dwell time” (DT). A low retention means that the DT must be taken into consideration, because exceeding the maximum allowed DT can result in unreliable read operations and ultimately, in data loss.

Since every read operation on a page in the block has a small detrimental effect on the pages in the same block, another important reliability consideration is the ability to reliably read a memory block after repeated read operations, referred to herein as the “read disturb” (RD) characteristic. As used herein, the RD characteristic of a memory device is defined as the maximum number of block reads after which a reliable read operation may still be performed. By way of example, some memory devices have a RD specification of about 10,000 block reads, after which the data in the block may not be reliably read.

In certain aspects of the subject technology, multiple memory blocks can be indexed according to three independent criteria. First, memory blocks can be indexed according to the number of invalid or garbage pages in the block (G). The value of G may vary with implementation, for example, in certain aspects G can range from 0 to N, wherein N represents the total number of pages in a given block (e.g., N=64 or 128). The value of G may be updated, for example, every time the host (e.g., host 110) declares pages of the block to be garbage.

Second, the memory blocks can be indexed according to the number of page reads (PRD), in the block, since the block was last written. Although the calculation of total page reads can vary with implementation, in some aspects each read performed on any page in a particular block will count as a page read. By way of example, if 100 pages in a block are read one time, and 50 pages are read two times, then the read count would be given by: PRD=100×1+50×2=200. Thus, in certain aspects the PRD can range from 0 (e.g., when the block has just been written), to maximum allowed value, (PRDmax). In certain aspects, PRDmax will be set at, or below, the RD specification value.

Third, memory blocks can be indexed based on a dwell time (DT), for example, relative to a time that the memory block was last written. The DT may be calculated using a timestamp that is saved at the time the block was last written. Earlier timestamp values can correspond to higher DT values, wherein the DT values can range from 0 (e.g., just after the block has been written), to a maximum allowed value (DTmax). In some implementations, DTmax is chosen such that the sum of DTmax and the maximum allowed time in the power-down state is equal to, or less than, the retention specification.

In some implementations, memory blocks are ranked using the foregoing criteria. In some examples, the higher the rank of the block, the earlier it is “recycled” by a garbage collection procedure, such as that described above with respect to FIGS. 2A and 2B. In one implementation, threshold values for PRDt and DTt, can be pre-defined, for PRD and DT indexes, respectively. Based on the values of PRD and DT, memory blocks in a flash memory array (e.g., flash memory array 130) can be associated with one of the following four rank group (RG) levels:

RG1: PRD≦PRDt, DT≦DTt

RG2: PRD>PRDt, DT≦DTt

RG3: PRD≦PRDt, DT>DTt

RG4: PRD>PRDt, DT>DTt

In certain aspects, PRDt can be set to a predetermined fraction of the maximum page reads. For example, PRDt may be set to ½ the maximum page reads. As such, in a flash device with RD of 10,000 block reads and 128 pages per block, PRDt can be set at ½×10,000×128=640,000. In other implementations, PRDt may be set more conservatively (e.g., ¼ of max page reads).

Similarly, the time duration value set for DTt may vary with implementation. For example, DTt could be set for a specific time duration, such as, 24 hours. However, in a more conservative configuration, DTt may be set to 12 hrs.

FIG. 3 illustrates the relative ranking of the four RG levels, according to one example. As illustrated, memory blocks in RG4 are scheduled to be recycled the earliest, followed by the highest ranking memory blocks of RG3 and RG2, respectively.

According to at least one implementation, the movement of memory blocks between RG levels occurs as follows. If PRD in a memory block exceeds PRDt, it is moved from RG1 to RG2 (or from RG3 to RG4). If the DT of the block exceeds DTt, it is moved from RG1 to RG3 (or from RG2 to RG4). When a block is newly written, it is moved from RG2, RG3 or RG4 to RG1 (or moved down in rank within RG1). Thus, more writes (e.g., from the host) can result in memory block movement to RG1, from the other of the RG levels. Additionally, fewer writes (or more reads) can result in movement from RG1 to RG2, RG3, or RG4.

Within every individual RG level, blocks may be ranked based on different criteria. In certain aspects, RG1 blocks can be ranked based on a number of pages containing garbage data. For example, blocks with a higher G have a higher ranking, as illustrated in FIG. 3.

In RG2, blocks can be ranked based on a value of PRD. That is, higher PRD values may be associated with a higher rank. In certain aspects, if any particular memory block in RG2 reaches PRD=PRDmax, the data in the memory block may be immediately relocated, irrespective of its current rank. That is, data from the memory block can be written into a different block, as shown in FIG. 3.

In RG3, blocks are ranked based on DT. That is, higher values of DT are associated with a higher rank. If any particular block in this group reaches DT=DTmax, the data in the memory block is immediately relocated, irrespective of its current rank, e.g., the data from this block is read and written into a different memory block, as indicated in FIG. 3.

In RG4, memory blocks can be ranked by the sum of PRD and DT (PRD+DT). If any memory block in this group reaches PRD=PRDmax or DT=DTmax, the data in the memory block is immediately relocated, irrespective of its current rank. That is, the data from the memory block is read and written into a different block, as indicated in FIG. 3.

In some examples, garbage collection can be performed on all flash memory blocks containing user data. However, in some implementations, blocks containing system data (e.g., tables, logs, etc.) will be exempted from garbage collection.

FIG. 4 illustrates another example ranking for RG levels. Specifically, in the example of FIG. 4, RG2 and RG3 are combined and RG1 blocks are ranked, within the RG1 level, based on a number of garbage pages. For example, in RG1 memory blocks with a greater number of “garbage” pages (e.g., higher G values) are given a higher ranking. As further illustrated in FIG. 4, RG2 and RG3 blocks are ranked by the value of a weighted sum of PRD and DT, given by: (k1×PRD+k2×DT), where k1 and k2 are weight coefficients, which can be chosen to prioritize either PRD or DT, as needed. Similarly, RG4 blocks are also ranked by the value of a weighted sum of PRD and DT, given by: (k3×PRD+k4×DT). By way of example, weighting coefficients k1, k2, k3 and k4 may be designated as various integer values based on the read disturb and retention characteristics of the corresponding memory device.

In certain aspects, weight coefficients k1, k2, k3 and k4 may be modified and/or updated at different times, for example, different phases of the memory lifecycle. In some aspects, coefficients may be adjusted based on a periodically measured bit error rate (BER), and once the BER reaches a certain threshold, the coefficients can be appropriately increased such that ranking priority is increased for the same PRD/DT. Additionally, PRDt and DTt can be decreased when a certain BER threshold is reached. In some examples, this may also increase ranking priority for the same PRD/DT by forcing memory blocks to a higher priority RG level.

If any particular block in RG2, RG3, or RG4 reaches PRD=PRDmax or DT=DTmax, the data in the block is relocated irrespective of its current rank, e.g., the data from this block is read and written into a different block.

It is understood that the foregoing ranking functions with respect to the RG1, RG2, RG3 and RG4 levels are examples of ranking functions based on read disturb and retention characteristics of a memory device. However, other ranking functions are possible, depending on the desired implementation.

In yet another aspect of the subject technology, values of PRDt and DTt may be chosen such that when a large amount of data is written to the corresponding memory device, a majority of blocks belonging to RG1, RG2, RG3 and RG4 are empty most of the time. In this example, since the blocks are being constantly re-written at high rate, there will be almost no blocks that require recycling based on DT or PRD (because after recycling a block both DT and PRD are reset to zero), and the maximum amount of newly erased space is recovered for every recycled block.

On the other hand, when a small amount of data is written to the memory device, the number of blocks in RG2, RG3, and RG4 will increase, and the blocks may be recycled based only on DT and/or PRD, thus keeping the drive operating reliably. In both examples, if PRD or DT in a given block reach their respective limits (e.g., PRDmax and DTmax), the corresponding memory block is immediately relocated to ensure that no excessive PRD or DT are allowed.

FIG. 5 illustrates an example of a write procedure 500 that can be performed to produce erased memory blocks out of memory blocks containing garbage data. The procedure 500 begins with step 502 in which a number of invalid pages across multiple data blocks is determined. Although any type of data stored to pages of a memory block may be designated as invalid, in certain implementations invalid data may correspond with data that has (or will be) updated and/or modified. In other implementations, invalid data may include corrupted or irrecoverable data stored to one or more pages in a memory device (e.g., flash memory array 130).

In step 504, a number of page reads for each of a multitude of data blocks is determined. As discussed above, a number of page reads (PRD) may be determined with respect to a specific duration of time (e.g., the number of page reads in a given time period), or with respect to a specific operation (e.g., the number of page reads may be tolled since a block was last written). Although the calculation used to determine the PRD may vary with implementation, in one example the PRD for a particular block may be calculated based on a number of times each page within the block is read. For example, if 100 pages in the block were read one time, and 50 pages were read two times, then read count would be given by: PRD=100×1+50×2=200.

In step 506, a dwell time for each of the multitude of data blocks is determined. The dwell time (DT) for a given data block may be tolled from various times, depending on implementation. As discussed above, the DT for a data block may be measured based on an amount of time that has elapsed since data was last written to the data block. In certain implementations, the DT for a particular data block may be resent (e.g., set to zero) each time the block is moved between RG levels.

In step 508, a data block is selected, from among the multitude of data blocks, for memory reclamation based on the number of invalid pages, the number of page reads and the dwell time of the selected data block. As discussed above, the selection of a data block for memory reclamation (e.g., for erasure) can be based on the associated RG level of the data block, which is based on measures the number of invalid pages, the number of page reads and the dwell time of the data block.

Thus, the subject technology provides an improved garbage collection process based on reliability considerations, e.g., based on PRD and DT. For example, in some implementations, as soon as a given block moves from RG1 to either RG2, (e.g., because PRD>PRDt), or to RG3 (because DT>DTt), reliability considerations prevail over the consideration to recover a maximum amount of newly erased space for every recycled block.

Those of skill in the art would appreciate that the various illustrative blocks, modules, elements, components, methods, and algorithms described herein may be implemented as electronic hardware, computer software, or combinations of both. To illustrate this interchangeability of hardware and software, various illustrative blocks, modules, elements, components, methods, and algorithms have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application. Various components and blocks may be arranged differently (e.g., arranged in a different order, or partitioned in a different way) all without departing from the scope of the subject technology.

It is understood that the specific order or hierarchy of steps in the processes disclosed is presented as an illustration of some exemplary approaches. Based upon design preferences and/or other considerations, it is understood that the specific order or hierarchy of steps in the processes can be rearranged. For example, in some implementations some of the steps can be performed simultaneously. Thus the accompanying method claims present elements of the various steps in a sample order, and are not meant to be limited to the specific order or hierarchy presented.

The previous description is provided to enable any person skilled in the art to practice the various aspects described herein. The previous description provides various examples of the subject technology, and the subject technology is not limited to these examples. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects. Thus, the claims are not intended to be limited to the aspects shown herein, but is to be accorded the full scope consistent with the language claims, wherein reference to an element in the singular is not intended to mean “one and only one” unless specifically so stated, but rather “one or more.” Unless specifically stated otherwise, the term “some” refers to one or more. Pronouns in the masculine (e.g., his) include the feminine and neuter gender (e.g., her and its) and vice versa. Headings and subheadings, if any, are used for convenience only and do not limit the invention.

The predicate words “configured to”, “operable to”, and “programmed to” do not imply any particular tangible or intangible modification of a subject, but, rather, are intended to be used interchangeably. For example, a processor configured to monitor and control an operation or a component may also mean the processor being programmed to monitor and control the operation or the processor being operable to monitor and control the operation. Likewise, a processor configured to execute code can be construed as a processor programmed to execute code or operable to execute code.

A phrase such as an “aspect” does not imply that such aspect is essential to the subject technology or that such aspect applies to all configurations of the subject technology. A disclosure relating to an aspect may apply to all configurations, or one or more configurations. An aspect may provide one or more examples. A phrase such as an aspect may refer to one or more aspects and vice versa. A phrase such as an “embodiment” does not imply that such embodiment is essential to the subject technology or that such embodiment applies to all configurations of the subject technology. A disclosure relating to an embodiment may apply to all embodiments, or one or more embodiments. An embodiment may provide one or more examples. A phrase such as an “embodiment” may refer to one or more embodiments and vice versa. A phrase such as a “configuration” does not imply that such configuration is essential to the subject technology or that such configuration applies to all configurations of the subject technology. A disclosure relating to a configuration may apply to all configurations, or one or more configurations. A configuration may provide one or more examples. A phrase such as a “configuration” may refer to one or more configurations and vice versa.

The word “exemplary” is used herein to mean “serving as an example or illustration.” Any aspect or design described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other aspects or designs.

Claims (16)

What is claimed is:
1. A method for managing memory operations in a storage device having a plurality of data blocks, the method comprising:
determining a number of invalid pages in each of the plurality of data blocks;
determining a number of page reads for each of the plurality of data blocks;
determining a dwell time for each of the plurality of data blocks;
associating the plurality of data blocks with a plurality of rank groups based on the number of page reads and the dwell time associated with each of the plurality of data blocks;
ranking each of the plurality of data blocks within the associated plurality of rank groups, wherein ranking within each of the plurality of rank groups is based on a set of criteria different from sets of criteria used for ranking within the other rank groups of the plurality of rank groups; and
selecting a data block, from among the plurality of data blocks, for memory reclamation based on the associated rank group and the ranking within the associated rank group of the selected data block.
2. The method of claim 1, wherein associating the plurality of data blocks comprises associating each of the plurality of data blocks with one of a first rank group, a second rank group, a third rank group, or a fourth rank group, based on the number of page reads and the dwell time associated with each of the plurality of data blocks.
3. The method of claim 1, wherein associating the plurality of data blocks comprises associating each of the plurality of data blocks with one of a second rank group, a third rank group or a fourth rank group based on a weighted sum of the number of page reads and the dwell time associated with each of the plurality of data blocks.
4. The method of claim 1, wherein:
the plurality of rank groups comprises a first rank group and a second rank group, ranking each of the plurality of data blocks comprises:
ranking each data block of the plurality of data blocks that is associated with the first rank group based on the number of invalid pages of the data block, and
ranking each data block of the plurality of data blocks that is associated with the second rank group based on at least one of the number of page reads of the data block or the dwell time of the data block.
5. The method of claim 1, wherein:
ranking each of the plurality of data blocks comprises ranking each data block of the plurality of data blocks that is associated with a second rank group based on a weighted sum of at least the number of page reads and the dwell time of the data block, and
weight coefficients associated with the weighted sum are adjusted based on a periodically measured bit error rate.
6. The method of claim 1, wherein:
the number of page reads for each of the plurality of data blocks is counted from a time a write operation was last performed on each of the plurality of data blocks, and
the dwell time for each of the plurality of data blocks is associated with a duration of time since a write operation was last performed on each of the plurality of data blocks.
7. A flash storage system comprising:
a memory;
a flash memory array comprising a plurality of flash memory blocks; and
a controller coupled to the memory and the flash memory array, wherein the controller is configured to perform operations for:
storing to the memory, a number of invalid pages associated with each of the plurality of data blocks;
storing to the memory, a number of page reads associated with each of the plurality of data blocks;
storing to the memory, a dwell time associated with each of the plurality of data blocks;
associating the plurality of data blocks with a plurality of rank groups based on the number of page reads and the dwell time associated with each of the plurality of data blocks;
ranking each of the plurality of data blocks within the associated plurality of rank groups, wherein ranking within each of the plurality of rank groups is based on a set of criteria different from sets of criteria used for ranking within the other rank groups of the plurality of rank groups; and
selecting a data block, from among the plurality of data blocks, for memory reclamation based on the associated rank group and the ranking within the associated rank group of the selected data block.
8. The flash storage system of claim 7, wherein associating the plurality of data blocks comprises associating each of the plurality of data blocks with one of a first rank group, a second rank group, a third rank group, or a fourth rank group based on the number of page reads and the dwell time associated with each of the plurality of data blocks.
9. The flash storage system of claim 7, wherein associating the plurality of data blocks comprises associating each of the plurality of data blocks with one of a second rank group, a third rank group, or a fourth rank group, based on a weighted sum of the number of page reads and the dwell time associated with each of the plurality of data blocks.
10. The flash storage system of claim 7, wherein:
the plurality of rank groups comprises a first rank group and a second rank group, ranking each of the plurality of data blocks comprises:
ranking each data block of the plurality of data blocks that is associated with the first rank group based on the number of invalid pages of the data block, and
ranking each data block of the plurality of data blocks that is associated with the second rank group based on at least one of the number of page reads of the data block or the dwell time of the data block.
11. The flash storage system of claim 7, wherein:
ranking each of the plurality of data blocks comprises ranking each data block of the plurality of data blocks that is associated with a second rank group based on a weighted sum of at least the number of page reads and the dwell time of the data block, and
weight coefficients associated with the weighted sum are adjusted based on a periodically measured bit error rate.
12. The flash storage system of claim 7, wherein:
the number of page reads for each of the plurality of data blocks is counted from a time a write operation was last performed on each of the plurality of data blocks, and
the dwell time for each of the plurality of data blocks is associated with a duration of time since a write operation was last performed on each of the plurality of data blocks.
13. The flash storage system of claim 7, wherein associating the plurality of data blocks is further based on a predetermined threshold value for the number of page reads and a predetermined threshold value for the dwell time.
14. A non-transitory computer-readable storage medium comprising instructions stored therein, which when executed by a processor, cause the processor to perform operations comprising:
determining a number of invalid pages in each of the plurality of data blocks;
determining a number of page reads for each of the plurality of data blocks;
determining a dwell time for each of the plurality of data blocks;
associating the plurality of data blocks with a plurality of rank groups based on the number of page reads and the dwell time associated with each of the plurality of data blocks;
ranking each of the plurality of data blocks within the associated plurality of rank groups, wherein ranking within each of the plurality of rank groups is based on a set of criteria different from sets of criteria used for ranking within the other rank groups of the plurality of rank groups; and
selecting a data block, from among the plurality of data blocks, for memory reclamation based on the associated rank group and the ranking within the associated rank group of the selected data block.
15. The non-transitory computer-readable storage medium of claim 14, wherein associating the plurality of data blocks comprises associating each of the plurality of data blocks with one of a first rank group, a second rank group, a third rank group, or a fourth rank group, based on the number of page reads and the dwell time associated with each of the plurality of data blocks.
16. The non-transitory computer-readable storage medium of claim 14, wherein:
the plurality of rank groups comprises a first rank group and a second rank group, ranking each of the plurality of data blocks comprises:
ranking each data block of the plurality of data blocks that is associated with the first rank group based on the number of invalid pages of the data block, and
ranking each data block of the plurality of data blocks that is associated with the second rank group based on at least one of the number of page reads of the data block or the dwell time of the data block.
US13/664,335 2011-11-18 2012-10-30 Optimized garbage collection algorithm to improve solid state drive reliability Active 2033-01-08 US9081663B2 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US201161561663P true 2011-11-18 2011-11-18
US13/664,335 US9081663B2 (en) 2011-11-18 2012-10-30 Optimized garbage collection algorithm to improve solid state drive reliability

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US13/664,335 US9081663B2 (en) 2011-11-18 2012-10-30 Optimized garbage collection algorithm to improve solid state drive reliability
US14/798,400 US9977736B2 (en) 2011-11-18 2015-07-13 Optimized garbage collection algorithm to improve solid state drive reliability

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US14/798,400 Continuation US9977736B2 (en) 2011-11-18 2015-07-13 Optimized garbage collection algorithm to improve solid state drive reliability

Publications (2)

Publication Number Publication Date
US20130132647A1 US20130132647A1 (en) 2013-05-23
US9081663B2 true US9081663B2 (en) 2015-07-14

Family

ID=48428065

Family Applications (2)

Application Number Title Priority Date Filing Date
US13/664,335 Active 2033-01-08 US9081663B2 (en) 2011-11-18 2012-10-30 Optimized garbage collection algorithm to improve solid state drive reliability
US14/798,400 Active US9977736B2 (en) 2011-11-18 2015-07-13 Optimized garbage collection algorithm to improve solid state drive reliability

Family Applications After (1)

Application Number Title Priority Date Filing Date
US14/798,400 Active US9977736B2 (en) 2011-11-18 2015-07-13 Optimized garbage collection algorithm to improve solid state drive reliability

Country Status (1)

Country Link
US (2) US9081663B2 (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9740609B1 (en) 2016-02-19 2017-08-22 International Business Machines Corporation Garbage collection techniques for a data storage system
US20180024751A1 (en) * 2016-07-19 2018-01-25 Western Digital Technologies, Inc. Metadata management on a storage device
US10095628B2 (en) 2015-09-29 2018-10-09 International Business Machines Corporation Considering a density of tracks to destage in groups of tracks to select groups of tracks to destage
US10120811B2 (en) 2015-09-29 2018-11-06 International Business Machines Corporation Considering a frequency of access to groups of tracks and density of the groups to select groups of tracks to destage
US10241918B2 (en) 2015-09-29 2019-03-26 International Business Machines Corporation Considering a frequency of access to groups of tracks to select groups of tracks to destage

Families Citing this family (36)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8533384B2 (en) 2007-12-27 2013-09-10 Sandisk Enterprise Ip Llc Flash memory controller garbage collection operations performed independently in multiple flash memory groups
US9081663B2 (en) 2011-11-18 2015-07-14 Stec, Inc. Optimized garbage collection algorithm to improve solid state drive reliability
US9699263B1 (en) 2012-08-17 2017-07-04 Sandisk Technologies Llc. Automatic read and write acceleration of data accessed by virtual machines
US9612948B2 (en) 2012-12-27 2017-04-04 Sandisk Technologies Llc Reads and writes between a contiguous data block and noncontiguous sets of logical address blocks in a persistent storage device
US9870830B1 (en) 2013-03-14 2018-01-16 Sandisk Technologies Llc Optimal multilevel sensing for reading data from a storage medium
US9015409B2 (en) * 2013-03-15 2015-04-21 Advanced Digital Broadcast Sa Apparatus and methods for prolonging service life of solid-state memory device in a digital video recorder
US9524235B1 (en) 2013-07-25 2016-12-20 Sandisk Technologies Llc Local hash value generation in non-volatile data storage systems
US9431113B2 (en) * 2013-08-07 2016-08-30 Sandisk Technologies Llc Data storage system with dynamic erase block grouping mechanism and method of operation thereof
US9639463B1 (en) * 2013-08-26 2017-05-02 Sandisk Technologies Llc Heuristic aware garbage collection scheme in storage systems
US9703816B2 (en) 2013-11-19 2017-07-11 Sandisk Technologies Llc Method and system for forward reference logging in a persistent datastore
US9520197B2 (en) 2013-11-22 2016-12-13 Sandisk Technologies Llc Adaptive erase of a storage device
US9520162B2 (en) 2013-11-27 2016-12-13 Sandisk Technologies Llc DIMM device controller supervisor
US9582058B2 (en) 2013-11-29 2017-02-28 Sandisk Technologies Llc Power inrush management of storage devices
US9703636B2 (en) 2014-03-01 2017-07-11 Sandisk Technologies Llc Firmware reversion trigger and control
US9448876B2 (en) 2014-03-19 2016-09-20 Sandisk Technologies Llc Fault detection and prediction in storage devices
US9454448B2 (en) 2014-03-19 2016-09-27 Sandisk Technologies Llc Fault testing in storage devices
US9626399B2 (en) 2014-03-31 2017-04-18 Sandisk Technologies Llc Conditional updates for reducing frequency of data modification operations
US9626400B2 (en) 2014-03-31 2017-04-18 Sandisk Technologies Llc Compaction of information in tiered data structure
US9697267B2 (en) 2014-04-03 2017-07-04 Sandisk Technologies Llc Methods and systems for performing efficient snapshots in tiered data structures
US10162748B2 (en) 2014-05-30 2018-12-25 Sandisk Technologies Llc Prioritizing garbage collection and block allocation based on I/O history for logical address regions
US10114557B2 (en) 2014-05-30 2018-10-30 Sandisk Technologies Llc Identification of hot regions to enhance performance and endurance of a non-volatile storage device
US10146448B2 (en) 2014-05-30 2018-12-04 Sandisk Technologies Llc Using history of I/O sequences to trigger cached read ahead in a non-volatile storage device
US9703491B2 (en) 2014-05-30 2017-07-11 Sandisk Technologies Llc Using history of unaligned writes to cache data and avoid read-modify-writes in a non-volatile storage device
US9653184B2 (en) 2014-06-16 2017-05-16 Sandisk Technologies Llc Non-volatile memory module with physical-to-physical address remapping
US9652381B2 (en) 2014-06-19 2017-05-16 Sandisk Technologies Llc Sub-block garbage collection
US20160232088A1 (en) * 2014-07-17 2016-08-11 Sandisk Enterprise Ip Llc Garbage Collection in Storage System with Distributed Processors
US20160070507A1 (en) * 2014-09-08 2016-03-10 Kabushiki Kaisha Toshiba Memory system and method of controlling memory device
US10025530B2 (en) 2014-09-29 2018-07-17 Western Digital Technologies, Inc. Optimized garbage collection for solid-state storage devices
CN104461920B (en) * 2014-12-09 2019-04-12 杭州华为数字技术有限公司 A kind of method and device of storing data
US20160179399A1 (en) * 2014-12-23 2016-06-23 Sandisk Technologies Inc. System and Method for Selecting Blocks for Garbage Collection Based on Block Health
KR20160090953A (en) 2015-01-22 2016-08-02 삼성전자주식회사 Storage device and sustained status accelerating method thereof
US9582192B2 (en) 2015-02-06 2017-02-28 Western Digital Technologies, Inc. Geometry aware block reclamation
CN106802867A (en) * 2015-11-25 2017-06-06 光宝电子(广州)有限公司 Solid state storage device and data programming method of same
US9690701B1 (en) * 2016-09-27 2017-06-27 International Business Machines Corporation Probabilistic, parallel collection of memory no longer in use
US10289548B1 (en) * 2017-04-28 2019-05-14 EMC IP Holding Company LLC Method and system for garbage collection in a storage system which balances wear-leveling and performance
US10303397B1 (en) * 2017-05-25 2019-05-28 Amazon Technologies, Inc. Read disturb handling in storage devices

Citations (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5532962A (en) * 1992-05-20 1996-07-02 Sandisk Corporation Soft errors handling in EEPROM devices
US7012835B2 (en) * 2003-10-03 2006-03-14 Sandisk Corporation Flash memory data correction and scrub techniques
US20080288814A1 (en) * 2007-05-16 2008-11-20 Jun Kitahara Apparatus for detecting and recovering from data destruction caused in an unaccessed memory cell by read, and method therefor
US20090172254A1 (en) * 2007-12-31 2009-07-02 Phison Electronics Corp. Method for preventing read-disturb happened in non-volatile memory and controller thereof
US20090193174A1 (en) * 2008-01-29 2009-07-30 Denali Software, Inc. Read disturbance management in a non-volatile memory system
US20100042772A1 (en) * 2008-08-14 2010-02-18 Addmm, Llc Method and apparatus for high reliability data storage and retrieval operations in multi-level flash cells
US20100217919A1 (en) * 2009-02-24 2010-08-26 Kabushiki Kaisha Toshiba Memory controller, semiconductor memory device and control method thereof
US7818525B1 (en) * 2009-08-12 2010-10-19 Texas Memory Systems, Inc. Efficient reduction of read disturb errors in NAND FLASH memory
US20110096601A1 (en) * 2009-10-28 2011-04-28 Gavens Lee M Non-Volatile Memory And Method With Accelerated Post-Write Read To Manage Errors
US20110225347A1 (en) * 2010-03-10 2011-09-15 Seagate Technology Llc Logical block storage in a storage device
US20120030506A1 (en) * 2010-07-30 2012-02-02 Apple Inc. Read disturb scorecard
US20120102260A1 (en) * 2010-10-26 2012-04-26 Hitachi, Ltd. Storage apparatus and data control method
US20120239990A1 (en) * 2009-05-12 2012-09-20 Stec, Inc. Flash storage device with read disturb mitigation
US20130091320A1 (en) * 2011-10-05 2013-04-11 Hitachi, Ltd. Storage system and storage method

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8285970B2 (en) * 2008-11-06 2012-10-09 Silicon Motion Inc. Method for managing a memory apparatus, and associated memory apparatus thereof
WO2011061724A1 (en) * 2009-11-23 2011-05-26 Amir Ban Memory controller and methods for enhancing write performance of a flash device
US8908431B2 (en) * 2010-02-17 2014-12-09 Samsung Electronics Co., Ltd. Control method of nonvolatile memory device
US9026716B2 (en) * 2010-05-12 2015-05-05 Western Digital Technologies, Inc. System and method for managing garbage collection in solid-state memory
US20120023144A1 (en) * 2010-07-21 2012-01-26 Seagate Technology Llc Managing Wear in Flash Memory
US9081663B2 (en) 2011-11-18 2015-07-14 Stec, Inc. Optimized garbage collection algorithm to improve solid state drive reliability
US8806111B2 (en) * 2011-12-20 2014-08-12 Fusion-Io, Inc. Apparatus, system, and method for backing data of a non-volatile storage device using a backing store

Patent Citations (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5532962A (en) * 1992-05-20 1996-07-02 Sandisk Corporation Soft errors handling in EEPROM devices
US7012835B2 (en) * 2003-10-03 2006-03-14 Sandisk Corporation Flash memory data correction and scrub techniques
US20080288814A1 (en) * 2007-05-16 2008-11-20 Jun Kitahara Apparatus for detecting and recovering from data destruction caused in an unaccessed memory cell by read, and method therefor
US20090172254A1 (en) * 2007-12-31 2009-07-02 Phison Electronics Corp. Method for preventing read-disturb happened in non-volatile memory and controller thereof
US20090193174A1 (en) * 2008-01-29 2009-07-30 Denali Software, Inc. Read disturbance management in a non-volatile memory system
US20100042772A1 (en) * 2008-08-14 2010-02-18 Addmm, Llc Method and apparatus for high reliability data storage and retrieval operations in multi-level flash cells
US20100217919A1 (en) * 2009-02-24 2010-08-26 Kabushiki Kaisha Toshiba Memory controller, semiconductor memory device and control method thereof
US20120239990A1 (en) * 2009-05-12 2012-09-20 Stec, Inc. Flash storage device with read disturb mitigation
US20110040932A1 (en) * 2009-08-12 2011-02-17 Texas Memory Systems, Inc. Efficient Reduction of Read Disturb Errors in NAND FLASH Memory
US7818525B1 (en) * 2009-08-12 2010-10-19 Texas Memory Systems, Inc. Efficient reduction of read disturb errors in NAND FLASH memory
US20110096601A1 (en) * 2009-10-28 2011-04-28 Gavens Lee M Non-Volatile Memory And Method With Accelerated Post-Write Read To Manage Errors
US20110225347A1 (en) * 2010-03-10 2011-09-15 Seagate Technology Llc Logical block storage in a storage device
US20120030506A1 (en) * 2010-07-30 2012-02-02 Apple Inc. Read disturb scorecard
US20120102260A1 (en) * 2010-10-26 2012-04-26 Hitachi, Ltd. Storage apparatus and data control method
US20130091320A1 (en) * 2011-10-05 2013-04-11 Hitachi, Ltd. Storage system and storage method

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
Bit Error Rate in NAND Flash Memories; Neal Mielke, Todd Marquart, Ning Wu, Jeff Kessenich, Hanmant Belgal, Eric Schares, Falgun Trivedi, Evan Goodness, and Leland R. Nevill; IEEE 2008. *
CATA: A Garbage Collection Scheme for Flash Memory File Systems by Longzhe Han, Yeonseung Ryu, and Keunsoo Yim; Myongji University and Samsung Advanced Institute of Technology; Springer Verlag Heidelberg 2006 p. 109. *
NAND Flash Solid State Storage for the Enterprise; An In-depth Look at Reliability; SNIA 2009. *
The Design and Implementation of a Log-Structured File System; Rosenblum and Ousterhout; ACM 1992. *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10095628B2 (en) 2015-09-29 2018-10-09 International Business Machines Corporation Considering a density of tracks to destage in groups of tracks to select groups of tracks to destage
US10120811B2 (en) 2015-09-29 2018-11-06 International Business Machines Corporation Considering a frequency of access to groups of tracks and density of the groups to select groups of tracks to destage
US10241918B2 (en) 2015-09-29 2019-03-26 International Business Machines Corporation Considering a frequency of access to groups of tracks to select groups of tracks to destage
US10275360B2 (en) 2015-09-29 2019-04-30 International Business Machines Corporation Considering a density of tracks to destage in groups of tracks to select groups of tracks to destage
US9740609B1 (en) 2016-02-19 2017-08-22 International Business Machines Corporation Garbage collection techniques for a data storage system
US20180024751A1 (en) * 2016-07-19 2018-01-25 Western Digital Technologies, Inc. Metadata management on a storage device

Also Published As

Publication number Publication date
US9977736B2 (en) 2018-05-22
US20150317247A1 (en) 2015-11-05
US20130132647A1 (en) 2013-05-23

Similar Documents

Publication Publication Date Title
US8966343B2 (en) Solid-state drive retention monitor using reference blocks
US8130553B2 (en) Systems and methods for low wear operation of solid state memory
US8904089B2 (en) Method for performing block management/Flash memory management, and associated memory device and controller thereof
US8438361B2 (en) Logical block storage in a storage device
US20150262659A1 (en) Setting operating parameters for memory cells based on wordline address and cycle information
US8650463B2 (en) Solid state drive and method of controlling an error thereof
US20100146191A1 (en) System and methods employing mock thresholds to generate actual reading thresholds in flash memory devices
US20140223084A1 (en) Memory system and related method of operation
US20130073798A1 (en) Flash memory device and data management method
US20120005558A1 (en) System and method for data recovery in multi-level cell memories
US8638602B1 (en) Background selection of voltage reference values for performing memory read operations
US20120047320A1 (en) Method and apparatus to interface semiconductor storage device and host to provide performance throttling of semiconductor storage device
US8656256B2 (en) Apparatus and method for multi-mode operation of a flash memory device
US8179718B2 (en) Memory device and memory programming method
US20090296466A1 (en) Memory device and memory programming method
US9442662B2 (en) Device and method for managing die groups
US9224483B2 (en) Nonvolatile memory device, system and programming method with dynamic verification mode selection
US8762629B2 (en) Data conditioning to improve flash memory reliability
US8203881B2 (en) Nonvolatile memory device, method of operating nonvolatile memory device and memory system including nonvolatile memory device
US9552288B2 (en) Multi-tiered memory with different metadata levels
US8799747B2 (en) Data hardening to compensate for loss of data retention characteristics in a non-volatile memory
US9368223B2 (en) Memory system and read reclaim method thereof
US9063844B2 (en) Non-volatile memory management system with time measure mechanism and method of operation thereof
US9135106B2 (en) Read level adjustment using soft information
US20130173856A1 (en) Data storage device, memory system, and computing system using nonvolatile memory device

Legal Events

Date Code Title Description
AS Assignment

Owner name: STEC, INC., CALIFORNIA

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:MELIK-MARTIROSIAN, ASHOT;REEL/FRAME:035374/0056

Effective date: 20150406

STCF Information on status: patent grant

Free format text: PATENTED CASE

AS Assignment

Owner name: HGST TECHNOLOGIES SANTA ANA, INC., CALIFORNIA

Free format text: CHANGE OF NAME;ASSIGNOR:STEC, INC.;REEL/FRAME:036042/0390

Effective date: 20131105

AS Assignment

Owner name: WESTERN DIGITAL TECHNOLOGIES, INC., CALIFORNIA

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:HGST TECHNOLOGIES SANTA ANA, INC.;REEL/FRAME:046174/0446

Effective date: 20180410

MAFP Maintenance fee payment

Free format text: PAYMENT OF MAINTENANCE FEE, 4TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: M1551); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

Year of fee payment: 4